BACKGROUND: Insulin assays are affected by varying degrees of interference from anti-insulin antibodies (IAs) and by cross-reactivity with recombinant insulin analogues. We evaluated the usefulness of the E170 insulin assay by assessing IA effects and cross-reactivity with 2 analogues. METHODS: Sera were obtained from 59 type 2 diabetes patients receiving continuous subcutaneous insulin infusion and 18 healthy controls. Insulin levels were determined using an E170 analyzer. To investigate the effects of IAs, we performed IA radioimmunoassays, and analyzed the differences between directly measured insulin (direct insulin) and polyethylene glycol (PEG)-treated insulins (free, IA-unbound; total, IA-bound and unbound insulin). We performed in-vitro cross-reactivity tests with insulin aspart and insulin glulisine. RESULTS: In IA-positive patients, E170 free insulin levels measured using the E170 analyzer were significantly lower than the direct insulin levels. The mean value of the direct/free insulin ratio and IA-bound insulin, which were calculated as the difference between total and free insulin, increased significantly as endogenous IA levels increased. The E170 insulin assay showed low cross-reactivities with both analogues (< 0.7%). CONCLUSIONS: IAs interfered with E170 insulin assay, and the extent of interference correlated with the IA levels, which may be attributable to the increase in IA-bound insulin, and not to an error in the assay. The E170 insulin assay may measure only endogenous insulin since cross-reactivity is low. Our results suggest that the measurement of free insulin after PEG pre-treatment could be useful for beta cell function assessment in diabetic patients undergoing insulin therapy.